Buying dips in stock indices, tested on 14 years.
Buy an index after a sharp one- or two-day drop, sell a few days later. We tested it on 10 indices from 2012 to 2026, and it kept working on the years we kept aside to check it.
The short answer
It works. Buying a stock index after a short, sharp drop and selling a few days later made money over 14 years, including the years we kept aside to check it. It is steady rather than spectacular: with a stop loss on every trade and 2% of the account risked per trade, it made 4.7% a year, and the account's deepest fall from a high point was 18.5%.
$10,000 traded with these rules
US30, NASDAQ 100, US500 and Japan 225, November 2012 to October 2026. Stop loss on every trade, 2% risk per trade, at most 3 trades open, trading costs included.
From $10,024 to $19,294 in 14 years, 459 trades.
The rules in plain words
- Only buy, never sell short.
- Only in an uptrend: the index is above its average price of the last 200 days.
- Buy after a sharp drop of a day or two: the RSI(2) indicator is below 10.
- Sell when the price closes above its average of the last 5 days, or after 10 days at the latest.
- Stop loss 3 times the average daily range (ATR 10) below the entry.
The idea is old and simple: stock indices tend to rise over time, and when one drops hard for a day or two inside an uptrend, it often bounces back within a few days. Larry Connors made it popular.
How we tested it
We let a computer trade the rules on 10 stock indices from 2012 to 2026, with trading costs included: 0.015% per side and 4% a year financing for every day a trade was held.
Then the most important step: we chose the indices using only 2012 to 2018, and checked them on 2019 to 2026, years the choice had never seen. That is your situation when you start a strategy today: it was chosen on the past, and you trade the future.
Five indices made money in 2012 to 2018. Four of them made money again in 2019 to 2026: US30, NASDAQ 100, US500 and Japan 225. Small changes to the rules (a bigger or smaller drop, 3 or 4 down days in a row) mostly worked too, a good sign that the result is real and not luck.
Index by index
Without a stop loss. "Won" = share of trades that made money; "average" = average gain per trade.
| Index | Trades | Won | Average |
|---|---|---|---|
| NASDAQ 100 | 121 | 73% | +0.49% |
| Japan 225 | 114 | 72% | +0.51% |
| US500 | 124 | 80% | +0.38% |
| US30 | 127 | 76% | +0.18% |
| Germany 40 | 123 | 70% | +0.09% |
| Hong Kong 50 | 73 | 64% | −0.12% |
| UK 100 | 100 | 59% | −0.06% |
Year by year
The $10,000 account above, stop loss on every trade, 2% risk per trade.
| Year | Result |
|---|---|
| 2012 (from Nov) | +0.5% |
| 2013 | +14.9% |
| 2014 | +6.9% |
| 2015 | +6.2% |
| 2016 | -0.7% |
| 2017 | +16.0% |
| 2018 | -8.4% |
| 2019 | -1.7% |
| 2020 | -5.5% |
| 2021 | +12.9% |
| 2022 | -5.4% |
| 2023 | +4.6% |
| 2024 | +4.6% |
| 2025 | +6.0% |
| 2026 (to Oct) | +21.5% |
Most years were up; the worst, 2018, lost 8.4%. In 2026 so far: +21.5%.
In the MetaTrader 5 Strategy Tester
We also ran the same rules in the MetaTrader 5 Strategy Tester on a broker's prices (4 indices, January 2020 to October 2026):
- 1% risk per trade: +15.8%, 254 trades, 72% won, profit factor 1.44, largest drawdown 6.5%.
- 3% risk per trade: +53.9%, largest drawdown 19.1%.
More risk per trade means more profit and deeper falls: the rules stay the same, only the size changes.
What it means for you
- The strategy is out of the market most of the time, so it runs well next to another strategy.
- It wins about 3 trades out of 4, and the stop loss keeps the losing ones to a planned size.
- It needs patience: a few trades a month across four indices.
Where Giza comes in
Giza Gold EA trades exactly these rules as its Index Dip part, on US500, NASDAQ 100, US30 and Japan 225, next to its gold breakouts, with a stop loss on every trade.
Every trade, in public.
Watch Giza Gold EA trade gold and stock indices, losses included, updated every 5 minutes.